Centralized Department of Defense Management of Cargo Shipped in Containers Would Save Millions and Improve Service

Published by the Government Accountability Office on 1977-11-08.

Below is a raw (and likely hideous) rendition of the original report. (PDF)

                         DOCUMENT RASOUB

04143 - [B3134407]
Centraliza2 Department )f Defense Banaqement of Cargo Shipped in
Containers Would Save Millions an4 Improve Service. LCD-77-227;
B-181'14. November 8, 1977. 10 pp. * appendix 42 pp.).
Report to Secretary, Department of Defense; by Robert G.
 IoRthwell (for Fred J. Shafer, Director, Logistics and
CoUmunicitions Div.).
Issue Area: Facilities and rateria.l Banagement: Federal
    Transportatioi of Things (704).
Contact: Logistics an. Communicatior.  Liv.
Budget FurctiGa: National Defense: Department of Defense -
    military (except procurement & contracts) (051).
Organization Concerned: Department cf the Navy: Military Sealift
    Command; Department of Defense: Bilitary Traffic Managa&ent
Congressional Relevance: House Comuittee on Armed Services;
    Senate Committee on Armed Services.
         The Military Traffic management Command and the
Military Sealife Command divide management responsibilities for
selecting conta:uaers for cargo going overseas, often preventing
the Department of Defense ({OD)from getting the lcYest
transportation costs. Both Commands sake adequate selections for
each portion of the movement, considering ;he information they
have and their responsibilities. Hovever, if shippers submitted
enough cargo information and responsibility and information were
centralized, better seLections could be made.
Findings/Conclusions: DOD could have saved about $12.9 millLon
of the more than $230 spent on overseas cargo shipments in
fiscal year 1976 though use of an effective central management
system and better use of up-to-date computer technology to
select the most economical shipping arrangement. Transportation
service to DOD shippers could also have been improved, reducing
the shipping tiaeso Less shi.pping time loers the inventory
costs which the Department estimated could be millions of
dollars each year. R.ecom.endations: The Secretary of Defense
should designate a central manager fcr carrier and contair4r
selections. The central manager should obtain detailed cargo
information from shippers and us5_ avail-able computer technology
to make the best selections for origin-to-destination shipments
of cargo sent overseas. (Author/SC)

 Centralized Department Of
 Defense Management Of
 Cargo Shipped In Containers
 Would Save Millions
 And Improve Service
The Military Traffic Management Command
and the Milita '¥ Sealift Command divide man-
agemgnt responsibilities for selecting contain-
ers for cargo going overseas. Each makes in-
depencent decisions based on different re-
sponsibilities and information. The Depart-
ment shculd:
     --Designate a central manager for carrier
       and container selections.
     --Use available computer technology to
       help make the best selections.

LCD.77-227                                   NOVEMBER 8, 1977
                               UNITED STATES GENERAL ACCOUNTING OFFICE
                                       WASHINGTON, D.C.   20548



         The Honorable
         The Secretary of Defense

                 Attention:         Assistant for Audit Reports
         Dear Mr. Secretary:

              This report discusses the practices used in shipment
         of containerized cargo overseas.

              As you know, section 236 of the Legislative Reorgan-
         ization Act of 1970 requires the head of a Federal agency
         to submit a written statement on actions taken on our
         recommendations to the Senate Committee on Governmental
         Affairs and the House Cm=nittee on Government Operations
         not later than 60 days after the date of the report and
         to the House and Senate Committees on Appropriations with
         the agency's first request for appropriations made more
         than 60 davs after the date of the report.

              Copies of this report are being furnished to the
         Secretaries of the Army, Navy, and Air Force.

                                                Sincerely yours,

                                                Fred J. Shafer
                                    AND IMPROVE SERVICE

           The Department of Defense spent over $200
           million during fiscal year 1976 to ship
           cargo overseas in ocean containers. Man-
           agement responsibilities for moving this
           cargo were divided between the Military
           Traffic Management Command and the Military
           Sealift Command, preventing Defense from
           getting the lowest transportation costs.
           Defense could have saved about $12.9 mil-
           lion through an effective central manage-
           ment system and better use of up-to-date
           computer technology to select the most
           economical shipping arrancment.

           Transportation service to Defense shippers
           could also have been improved, reducing
           shipping times. Less shipping time lowers
           inventory costs, which the Department es-
           timated could be millions of dollars each

           Today, about 77 percent of the cargo sent
           overseas goes in ocean containers. Ship-
           pers load cargo into containers inland
           and the Traffic Management Command ar-
           ranges for delivery at overseas destina-
           tions without rehandling the cargo. This
           is an origin-to-final-destination trans-
           portation system.

           The container and carrier selection system
           is not achieving its objective of moving
           cargo at the lowest transportation cost.
           The two Commands independently decide what
           to do, based on their different responsi-
           bilities and different (and in the case of
           the Traffic Management Command, insuffi-
           cien%) information. This split prevents

1Tr Sht. Upon removal. the report
cover date should be noted heron.   i                    LCD-77-2Z7
the best container from being selected and
causes excess transportation costs.
Excess transportation costs
Both Commands make adequate selections for
each portion of the mevement, considering
the information they have and their respon-
sibilities. However. if shippers submitted
enough cargo information and responsibility
and information were at a central point,
better selections could be made.

GAO studied 304 container requests from
Department of Defense shippers from the
east and west coasts. The Commancs se-
lected 815 containers costing $1,015,480
to move the cargo.
Using a simulated central management system,
GAO selected containers on tFh same re-
quests. Only 780 containers would have
been required to move the -ame cargo, re-
ducing transportation costs by $65,098.
GAO changed container selections:

--To adjust the cointai.er size.

--To use a different carrier.

-- To use a combination of container sizes.
-- To use a combination of container sizes
   between different carriers.
For example, the Commands shipped 2,520
cubic feet of general cargo weighing 84,200
pounds in two large containers for $3,210.
The same cargo could have been shipped
in three small containers, saving $986.

The Commands are responsible for providing
prompt service to shippers. The fragmented
management, however, requires an additional

             2 to 3 days for container selections and book-
             ings. Defense estimated that reducing these
             delays could save millions of dollars annually
             by decreasing inventory.
             Neither Command has a coot performance measure.
             ment system. By adopting centralized manage-
             ment and by determining what the best container
             is, Defense could develop a cost performance
             measure. Actual bookings could be compared
             to the ideal, not only to measure coct perfor-
             mance but also to determine why the best selec-
             tion was not made.

             The Secretary of Defense should designate a
             central manager for ca-rier and container
             selections. The central manager should ob-
             tain detailed cargo information from shippers
             and use available computer technology to make
             the best selections for origin-to-destination
             shipments of cargo sent overseas.
             AGENCY COMMENTS

             Commenting on GAO's report, the Department
             of Defense agreed that more extensive use
             of computer technology might help to make
             the most cost-effective selection of con-
             tainers. DOD added that it is carefully
             considering designating a central manager.

Tear Sheet                        1iii



       1   INTRODUCTION                                   1
               Fragmented management system               3
               Excess transportation costs result         4
               Potential transportation service
                 improvements                             6
               Potential cost performance measure-
                 ment system                              7
             COMMENTS AND OUR EVALUA'ION                  8
               Conclusions                                8
               Recommendations                            8
               Agency comments and our evaluation         8
      5    SCOPE OF REVIEW                               10

      I    Letter dated October 7, 1977, from the
             Principal Deputy Assistant Secretary
             of Defense (Manpower, Reserve Affairs
             and Logistics)                              11
CONUS      continental United States
DOD        Department of Defense
GAO        General Accounting Office
MSC        Military Sealift Command
MTMC       Military Traffic Management Command
                          CHAPTER 1

     In 1966 almost no Department of Defense (DOD) cargo was
containerized. Break-bulk shipping was the only surface
transportation system available to move cargo overseas. The
Military Traffic Management Command (MTMC) and the Military
Sealift Command (MSC) divided management responsibilities
along the lines of inland versus ocean movements. For ship-
ments originating inland, MTMC managed the movement to the
continental United States (CONUS) port, entered the shipment
into the terminal system, and held the cargo until enough
had accumulated for MSC booking. MSC then managed the cargo
movement from the CONUS port to the overseas port. Theater
commanders arranged delivery from overseas ports. in this
mode of shipping, MTMC managed carrier selection for land
movements and MSC managed ocean carrier selection.
     Container shipping began to replace break-bulk shipping
as the primary surface mode for overseas cargo movement dur-
ing the mid-1960s. Toda,, about 77 percent of the cargo is
shipped in containers. Shippers now load cargo into con-
tainers at the inland origin and arrange for delivery at the
overseas destination without rehandling the cargo. This
is a true systems-oLiented transportation mode. Container
management differs from break-bulk management because there
is no clear distinction between land and ocean movement.
Containerized cargo does not go through military ocean termi-
nals, buc is handled as a single move through commercial
ports in CONUS and overseas. However, MTMC and MSC manage-
ment responsibilities remain the same as they were in the
break-bulk system.
     In the current carrier and container selection system,
shippers request containers from MTMC based on cargo on
hand or projections of anticipated cargo shipments. These
requests give MTMC information on the shipment's origin;
destination; commodity; and, in some cases, weight and cube.
MTMC reviews these container requests and, without change,
requests ocean carrier booking from MSC. The booking re-
quests give MSC information on the number and size of con-
tainers needed, ports of embarkation and debarkation, and
type of inland drayage service. MSC makes a low cost
analysis of the available ocean carriers that will satisfy
the MTMC booking request. MSC books the shipment and noti-
fies MTMC, which in turn notifies the shipper.

     DOD recognized the difference between break-bulk and
container management, and in 1971 it proposed consolidating
MTMC and MSC, so that the total surface transportation sys-
tem, from origin to destination, could be centrally managed.
This consolidation also would have reduced the time required
to coordinate shipment requests between the two agencies;
DOD estimated that this could result in annual inventory
savings of more than $1 million. For various reasons, the
Congress did not approve the consolidation.

     In this report we describe how container selection,
which is now separately performed by each organization, can
be centrally managed for improved econrmy and efficiency
even though MTMC and MSC continue as separate commands. In
a separate, but related report, now being prepared, we will
address the need for more progress by DOD in developing an
effective container system for use in a contingency.

                            CHAPTER 2
       The container end carrier selection
 ing its objective of moving cargo          system is not achiev-
 and MSC make independent decisions at  the lowest cost. MTMC
 ment responsibilities and different based on different manage-
 receives insufficient information     infcrmation. Also, MTMC
                                    from shippers about the
 cargo to be shipped. This fragmertation
 of the optimal container and results       prevents selection
 costs.                                 in excess transportation

      DOD's container shipping costs
 fiscal year 1976. Through a central were $201 million in
                                       decisionmaking system
 for container and carrier selections,
                                        we estimate that over-
 seas shipping costs could be reduced
 annually.                             about $12., million

      As discussed previously, MTMC and
                                         MSC divide management
responsibilities for overseas cargo
the number and size of containers    movement.    MTMC selects
                                   needed baseo on shipper
requirements, then selects the type
service and the port of embarkation. of inland transportation
                                        MSC manages the carrier
selection and bookings based on MTM:
for ocean transportation or a combination         which can be
transportation. Neither MTMC nor            of  ocean  and land
                                   MSC  manages  the  entire

     MTMC is responsible for selecting
                                        the lowest overall
cost route for overseas cargo. MTMC
overseas inland transportation cost develops CONUS and
vides MTMC with ocean transportation information. MSC pro-
                                      cost data and other
ocean carrier charges.
      Shippers do not always send cargo
                                         weight, cube, and
dimension information.   Frequently, they request only a
specific number of small or large
give MTMC container and carrier    containers. MSC could
                                 availability  information,
but it does not do so because MSC
                                   is responsible for select-
ing the ocean carrie:. MTMC could
                                    challenge container re-
quests from shippers and select the
                                     optimal container and
carrier, but without adequate, information
view requests and submit them to MSC        it can only re-
                                      for ocean booking.

     MSC makes a cost analysis of available carriers that
will satisfy the MTMC request and books the shipment with
an ocean carrier. MSC officials said their responsibility
is to provide the service requested by MTMC, if that service
is available within the required sailing dates.

     MTMC and MSC officials agree that, because management
responsibilities age segmented, decisionmaking is fragmented.
They also agree that eliminating this fragmentation could re-
duce transportation costs. They believe that their primary
responsibility is to provide transportation services to
shippers at the lowest possible cost.

     MTMC and MSC generally make appropriate decisions, con-
sidering the information in their possession and their re-
sponsibilities. However, if all information were available
at a central point, optimal selections could be made by a
responsible manager. The computer used by MTMC is capable
of selecting the most cost-effective container.

     To determine the overall benefit of a central decision
point, we studied 153 general cargo shipments from the west
coast to Japan and 151 such shipments from the east coast
to Germany. We simulated a central management system and
developed a mathematical model, using all pertinent available
information, to determine the optimal container selection.
This information came from MSC and MTMC records. The model
selections considered only containers that were available
when actual bookings were made.
     A comparison of the model and actual selections indi-
cates the potential for savings through centralized decision-
making. Actual transportation costs for the 304 container
requests studied were $1,015,480. Had the requests been
managed in a central system, 124 requests would have been
changed and costs thereby reduced $65,098 (or 6.4 percent).

    Container requests were changed for the following reasons:
    -- To adjust the container numbers and sizes (from large
       to small or small to large).
    -- To use a different carrier.

    -- To use a mix of container sizes.

    -- To use a mix of container sizes between different

     The following examples illustrate container selection

Adjustment of container
number and sizes

     The Air Force Water Port Logistics Office Distribution
Center, Bayonne, New Jersey, requested two 40-foot containers
to ship 2,520 cubic feet of general cargo weighing 84,200
pounds to the Kastel Air Station, Germany. MTMC and MSC
reviewed the request and selected and booked two United States
Lines 40-foot containers for $3,210. Based on available in-
formation and according to the quantitative selection model,
that cargo could have been moved in three 20-foot United
States Lines containers, saving $987.

Use a different carrier
     Each week Sharpe Army Depot requested one large con-
tainer to ship general cargo to Sagami, Japan. MTMC and
MSC selected and booked one Sealand 40-foot container at
$1,929, with Sealand responsible for CONUS drayage. The
model, however, selected one United States Lines 40-foot
container, saving $32.

Use a mix of container sizes

     The General Services Administration's Supply Distribu-
tion Facility in Auburn, Washington, requested five large
containers to ship 112,400 pounds--9,000 cubic feet--of
general cargo to the U.S. Naval Supply Depot, Yokosuka,
Japan. MTMC and MSC selected and booked five Sealand 40-
foot containers at $9,375. The model selected one Sealand
40-foot and four Sealand 35-foot containers to move the
cargo, saving $948.

Use a mix of container sizes
between different ocean carriers

     The Army and Air Force Exchange Service requested 25
larqe or 42 small containers to move 1,118,250 pounds--
33,950 cubic feet on 750 pallets--of beverages from Long
Island City, New York, to the Support Center, Roedgeners,
Geissen, Germany. MTMC and MSC selected and booked 25 Sea-
land 35-foot containe:r as the best alternative. The total
cost was $35,925. The model considered possible alternatives
and selected I Sealand 35-foot and 40 United States Lines
20-foot containers, saving $608.

                            CHAPTER 3

     Centralized management of container and carrier selection
offers additional benefits. The system would be more respon-
sive to the transportation needs of DOD shippers because de-
lays caused by coordination between MTMC and MSC would be
avoided. Also, the quantitative selection model would provide
a cost performance measurement system that could identify sys-
tem weaknesses.

     MTMC and MSC are responsible for providing timely trans-
portation service to DOD shippers. Yet the coordination
process between the two agencies causes delays of up to 3 days
in container selections and bookings. DOD, in considering the
1971 proposed consolidation, felt that reducing these booking
delays could result in substantial pipeline inventory savings.

     To determine the effect of centralized management and
container request changes, we discussed our findings with
DOD shippers and commercial vendors on the east and west
coasts. All shippers agreed that central decisionmaking
would reduce the time required to receive bookings and would
help their operations because they could coordinate their
shipping requirements at one point. All but one of the
shippers and vendors felt that shipments moving in two dif-
ferent size containers and between two different carriers
uould not affect their shipping procedures.

     The New Cumberland Army Depot and the Mechanicsburg
Defense Depot require procedures different from those used
for other DOD shippers. Currently, these depots ship such
large volumes of cargo that MTMC and MSC cannot handle
individual requests. As a result, both depots request
containers in large allocations 3 to 5 weeks in advance.
These allocations, referred to as container pools, include
a mixture of container sizes belonging to various ocean
carriers. For example, Mechanicsburg may require 130 con-
tainers a week, but the cargo to be shipped is not known
when the request is made. MTMC and MSC merely satisfy the
request. These two shippers, however, said that a cen-
tralized management system would benefit their operations
because they could deal with one agency instead of two.

The quantative selection model could also help in selecting
the best containers from the container pool when all ship-
ment information is available and the cargo is about to be


     MTMC and MSC do not know how cost effectively they are
providing transportation services. MTMC uses container utili-
zation reports to measure its performance.  However, these
utilization reports, which are prepared after the cargo is
shipped, only indirectly indicate effectiveness at making
container selection. What MTMC needs is cost standards it
could apply to choose the most cost-effective arrangement
of available containers, carriers, and routes before the
shipment is made. Since MSC is only providing a service to
MTMC and shippers, it does not have a measurement system.

     A central decisionmaking system could develop a viable
cost performance measurement. By selecting the optimal
carrier and container, a cost performance objective is
developed. The actual costs could be compared to the esti-
mates to measure performance and indicate whether the opti-
mal selection was made. This would provide a basis for
investigating deviations and taking necessary corrective

                            CHAPTER 4

     Within the constraints of the cargo information they re-
ceive and their current responsibilities, MTMC and MSC gen-
erally make good carrier and container selections to ship DOD
cargo overseas. A centralized decisionmaker, however, could
make optimal selections which would save transportation costs,
improve transportation service, and develop criteria for a
cost performance measurement system.

     MTMC is the agency responsible for selecting the lowest
overall cost route for overseas cargo movement. MTMC has
the computer technology available to make the most cost-
effective container selection.


     We recommend that the Secretary of Defense designate a
central manager for carrier and container selections on con-
tainerized shipments. Although we have not examined all of
the command relationships and prerogatives of MTMC and MSC,
it is apparent that MTMC has a broader traffic management
responsibility-, the computer capability and expertise, and
greater responsibilities and relationships with shippers
and shipping data. Therefore, within the specific command
relationships covered by this study, we believe MTMC is the
logical choice for the centralized management responsibility.

     We also recommend that the Commanding General, MTMC,
(1) require shippers to provide the cargo data needed to
make the most cost-effective selections of containers and
(2) use available quantitative techniques and computer
technology in making the selections.

     We furnished a draft of this report to DOD for review.
Its comments are included as appendix I.

     DOD agreed that, within the constraints of data re-
ceived and their current responsibilities, MTMC and MSC
make good carrier and container selections to ship DOD

cargo overseas. DOD also agreed that
                                       mori extensive use
of applicable computer technology
                                   might help to make the
most cost-effective selection of containers
excellent tool for managers. However,         and can be an
timeliness and accuracy of weight       DOD said that the
shippers was a key to the entire issue.cube data  provided by
to potential problems of con:aineL        DOD  also alluded
     We agree that accurate weight and
                                        cube information is
necessary to determine container requirements.
many large shippers, such as vendors             However,
                                      of beverages, who have
standard load configurations can and
weight and cube data when requesting  do provide accurate
opinion, other DOD shippers also coulda booking. In our
                                        provide estimated
weight and cube data based on an experience
commodities.                                 factor for their

     DOD contends that the computer model
for nonavailability of containers          made nc allowance
at a given time, and that availability a  particular carrier
dynamic data element and should not    of  containers is a
                                    be considered a constant.
     We recognize that the availability
a dynamic data element. In the tests     of containers is
we only considered containers that     described on page
                                   carriers stated were 4,
available. Likewise, the computer
                                   model considers avail-
ability of containers to be a variable
dat element.                            and not a constant

     DOD said that the matter of designating
for carrier and containcr selection          a central manager
that the command relationships and was being considered and
and MSC were among the factors to responsibilities of MTMC
                                  be taken into account.

                          CHAPTER 5
                       SCOPE OF REVIEW
     We evaluated DOD's container and carrier selection sys-
tem for moving cargo overseas at the Military Traffic
agement Command, Headquarters, Western and Eastern Areas,
and the Military F-:alift Command, Pacific and Atlantic.
also visited and Interviewed DOD shipper services, commer-We
cial vendors, and commercial ocean carriers on the west
east coasts.                                             and

     We interviewed DOD, MTMC, and
and evaluated procedures; examined KSC officials; reviewed
                                   files and records; and
studied pertinent legislative actions.

APPENDIX          I                                                    APPENDIX I

                          ASSISTANT SECRETARY OF DEFENSE
                                 WASHINGTON, D C   20301

                                                             October   7,   1977

       Mr. Fred J. Shafer
       Logistics and Communications Division
       U.S. General Accounting Office
       Washington, D. C. 20548

       Dp-r Mr. Shafer:

       This is in response to your letter of July
                                                  18 to the Secretary of Defense
       tra Ismitting copies of your draft report
                                                 on "Centralized Management
       for Cargo Container Selections Would
                                               Reduce Transportation Costs and
       Improve Service" (OSD Case #4669).

       We agree with the conclusion that within
       received and their current responsibilitie the constraints of the data
                                                   s the Military Traffic Manage-
      ment Command (MTMC) and the Military
                                                   Sealift Command (MSC) make
      valid carrier and container selections
                                               to ship DoD cargo overseas. We
      also agree that more extensive use of applicable
                                                            computer technology
      may very well assist in optimizing the
                                                cost effecihve selection of con-
      tainers. However, key to the entire issue
                                                     is the
      of data received from shippers. Th.: assumption timeliness and accuracy
      cube data are available at the time containers         that exact weight and
                                                         are requested is overly
      optimistic. More often than not, large
                                                DoD shipping activities request
      containers prior to actual avaiiability of
                                                 cargo, based on experience and
      historical cargo generation patterns,
                                              so that the container will be ready
      when the cargo is offered. This practice
      required warehouse space, and personnel. reduces pipeline time inventories,
                                                      These activities do not operate
      under an allocation systemi but are provided
                                                       containers specifically
      requested based on need. Further, there
      pools available contingent on commercial may or may not be container
                                                   demand for containers.
     Although New Cumberland Army Depot
                                               and Mechanicsburg Defense
     Depot produce a significant volLme of
                                              shipr.lents, they are not so great
     that MTMC/MSC cannot handle processing
                                                    individual requests.
     We agree that the optimization techniques
                                               of linear programming can be
     excellent tools for I ,anagers. Care must
                                               be taken to insure appropriate
     constraints al-e included however. Otherwise
                                                   the optimization process,
 APPENDIX I                                                                APPENDIX I

   although correct in itself, can produce unrealistic results.
                                                                  For example,
   your computer model made no allowance for non-availability
                                                                  of containers
   front a particular carrier at a given time. Availability of
                                                                containers is
   a dynamic data element and should not be considered a constant.

   Ycur recommendation that the Secretary of Defense designate
                                                                  a central
   manager for carrier and container selections on containerized
   is being carefully considered. We are currently in the process
                                                                    of analyzing
   revisions to the single manager charters. While eventual
                                                              designation of
   a central manager tor container selection may serve to increase
   and improve service, our review of MTMC and MSC command
   and responsibilities must also be an element in such a decision.


                                                   ROBERT B. PIR(, JR.
                                           Principal Deputy Assistant Seoretary
                                                   of Defenlse (MRA&L)